基于熵权的集对分析模型在水质综合评价中的应用

针对水质评价中各评价指标的不确定性，将集对分析理论应用于水质综合评价中。先通过计算评价样本与评价指标之间的联系度对样本做初步分类，再对样本做进一步的同一、差异、对立的集对分析以判断评价样本的等级。在确定各评价指标的权重时，将信息论中的熵值理论引入该模型，运用信息熵所反映实测数据的效用值计算各评价指标的权重，使权重的分配具有一定的理论依据。最后运用蒙特卡罗法构造算例，讨论了实测过程中的随机观测误差对评价结果的影响。经实例分析，通过与综合评判法、属性识别法和模糊物元法的比较表明，基于熵权的集对分析模型的评价结果合理、客观。

Application of set pair analysis model based on entropy weight to comprehensive evaluation of water quality

In view of the uncertainty of evaluation indicators for water quality the theory of set pair analysis is adopted. The samples are preliminarily classified by calculating the connection degree between the samples and evaluation indicators. Then the samples are further ranked through the identity, difference and antagonism set pair analysis. The weights of evaluation factors in the model are obtained from the avail value of data reflecting the information entropy, by which the weight allocation is more reasonable. Besides, the influence of stochastic observation error on the evaluation result is investigated using Monte Carlo method. The application result of this model to an example is compared with those obtained from integrated evaluating methods, the attribute recognition method and fuzzy matter_element model. It shows that the proposed method is practical and reasonable.